Bridging or Breaking: Impact of Intergroup Interactions on Religious Polarization (2402.11895v3)

Abstract: While exposure to diverse viewpoints may reduce polarization, it can also have a backfire effect and exacerbate polarization when the discussion is adversarial. Here, we examine the question whether intergroup interactions around important events affect polarization between majority and minority groups in social networks. We compile data on the religious identity of nearly 700,000 Indian Twitter users engaging in COVID-19-related discourse during 2020. We introduce a new measure for an individual's group conformity based on contextualized embeddings of tweet text, which helps us assess polarization between religious groups. We then use a meta-learning framework to examine heterogeneous treatment effects of intergroup interactions on an individual's group conformity in the light of communal, political, and socio-economic events. We find that for political and social events, intergroup interactions reduce polarization. This decline is weaker for individuals at the extreme who already exhibit high conformity to their group. In contrast, during communal events, intergroup interactions can increase group conformity. Finally, we decompose the differential effects across religious groups in terms of emotions and topics of discussion. The results show that the dynamics of religious polarization are sensitive to the context and have important implications for understanding the role of intergroup interactions.

• The paper finds that intergroup interactions generally reduce religious polarization, with diminished effects among individuals with high baseline conformity.
• The paper introduces a novel Group Conformity Score derived from contextualized tweet embeddings to quantify semantic divergence in religious discourse.
• The study reveals that intergroup engagements can either bridge or exacerbate divides depending on event context, such as political neutrality versus communal tension.

Analyzing the Impact of Intergroup Interactions on Religious Polarization in the Context of COVID-19 on Twitter

Introduction

The dynamics of religious polarization amidst crises have garnered significant attention due to their potential to influence social cohesion and collective responses to emergencies. This paper investigates the role of intergroup interactions between majority and minority religious groups on social media platforms, specifically Twitter, in the context of the COVID-19 pandemic. Utilizing a dataset comprising nearly 700,000 Indian Twitter users engaging in discourse related to COVID-19 during 2020, the paper explores how these interactions impact group conformity and, by extension, religious polarization.

Methodology

To quantify religious polarization, the authors introduce a novel metric based on contextualized embeddings of tweet text, termed Group Conformity Score (GCS), which assesses the semantic divergence in topics and emotions between religious groups. By leveraging a meta-learning framework, the paper examines the heterogeneous treatment effects of intergroup interactions across various communal, political, and socio-economic events. Specifically, the paper analyzes the changes in group conformity pre- and post-interaction to ascertain whether intergroup exchanges bridge or exacerbate religious divides.

Results

The findings reveal that intergroup interactions generally reduce religious polarization, particularly in the context of political and social events. However, this depolarizing effect lessens for individuals exhibiting high baseline levels of group conformity. Conversely, during communal events, intergroup interactions tend to increase group conformity among minority group members, suggesting a polarization effect in scenarios directly implicating the minority community. Additionally, the decomposition of treatment effects across religious groups highlights that changes in emotional expressions and discussion topics significantly contribute to the observed variations in the impact of intergroup interactions on polarization.

Discussion

The paper provides a nuanced understanding of the conditions under which intergroup interactions can either mitigate or worsen religious polarization on social media. It highlights the importance of context, showing that while interactions can foster understanding in neutral settings, they might reinforce divisions during communally charged events. The differential effects observed between majority and minority groups underscore the complex interplay between social identity, informational exposure, and event-specific sensitivities.

Implications and Future Directions

This paper contributes to the broader discourse on polarization, social media dynamics, and intergroup relations by providing empirical evidence on the contextual factors that modulate the outcomes of intergroup interactions. The findings suggest potential strategies for social media platforms and policymakers to encourage more constructive intergroup engagements, such as promoting inclusive dialogue in politically neutral or supportive contexts while being mindful of the risks during communal tensions.

Future research may extend this work by examining the long-term effects of intergroup interactions on social cohesion and exploring mechanisms to counteract polarization in digital spaces. Additionally, investigating other dimensions of identity, such as political affiliation or ethnicity, could yield further insights into the multifaceted nature of polarization in online communities.